(1S,2R/1R,2S)-cis-cyclopentyl PNAs (cpPNAs) as constrained PNA analogues: Synthesis and evaluation of aeg-cpPNA chimera and stereopreferences in hybridization with DNA/RNA

Article abstract of DOI:10.1021/jo049442+

Conformationally constrained chiral PNA analogues were designed on the basis of stereospecific imposition of a 1,2-cis-cyclopentyl moiety on an aminoethyl segment of aegPNA. It is known that the cyclopentane ring is a relatively flexible system in which the characteristic puckering dictates the pseudoaxial/pseudoequatorial dispositions of substituents. Hence, favorable torsional adjustments are possible to attain the necessary hybridization- competent conformations when the moiety is imposed on the conventional PNA backbone. The synthesis of the enantiomerically pure 1,2-cis-cyclopentyl PNA monomers (10a and 10b) was achieved by stereoselective enzymatic hydrolysis of a key intermediate ester 2. The chiral (1S,2R/1R,2S)-aminocyclopentylglycyl thymine monomers were incorporated into PNA oligomers at defined positions and through the entire sequence. Hybridization studies with complementary DNA and RNA sequences using UV-T_m measurements indicate that aeg-cpPNA. chimera form thermally more stable complexes than aegPNA with stereochemistry-dependent selective binding of cDNA/RNA. Differential gel shift retardation was observed on hybridization of aeg-cpPNAs with complementary DNA.

Full text of DOI:10.1021/jo049442+

(1S,2R/1R,2S)-cis-Cyclop en tyl P NAs (cpP NAs) a s Con str a in ed P NA
An a logu es: Syn th esis a n d Eva lu a tion of a eg-cpP NA Ch im er a a n d
Ster eop r efer en ces in Hybr id iza tion w ith DNA/RNA
T. Govindaraju, Vaijayanti A. Kumar,* and Krishna N. Ganesh*
Division of Organic Chemistry (Synthesis), National Chemical Laboratory, Pune, 411 008, India
kng@ems.ncl.res.in
Received April 5, 2004
Conformationally constrained chiral PNA analogues were designed on the basis of stereospecific
imposition of a 1,2-cis-cyclopentyl moiety on an aminoethyl segment of aegPNA. It is known that
the cyclopentane ring is a relatively flexible system in which the characteristic puckering dictates
the pseudoaxial/pseudoequatorial dispositions of substituents. Hence, favorable torsional adjust-
ments are possible to attain the necessary hybridization-competent conformations when the moiety
is imposed on the conventional PNA backbone. The synthesis of the enantiomerically pure 1,2-cis-
cyclopentyl PNA monomers (10a and 10b) was achieved by stereoselective enzymatic hydrolysis of
a key intermediate ester 2. The chiral (1S,2R/1R,2S)-aminocyclopentylglycyl thymine monomers
were incorporated into PNA oligomers at defined positions and through the entire sequence.
Hybridization studies with complementary DNA and RNA sequences using UV-Tm measurements
indicate that aeg-cpPNA chimera form thermally more stable complexes than aegPNA with
stereochemistry-dependent selective binding of cDNA/RNA. Differential gel shift retardation was
observed on hybridization of aeg-cpPNAs with complementary DNA.
In tr od u ction
In the area of modified nucleic acids and oligonucle-
otide analogues, several chemical modifications have
been reported to date for applications in antisense
therapeutics.1 Among these, peptide nucleic acids (PNAs),
composed of N-(2-aminoethyl)glycine units to which
natural nucleobases are attached via a methylenecar-
bonyl linker (Figure 1, PNA II), are found to mimic
many of the oligonucleotide properties.2 PNA binds to
complementary DNA and RNA to form duplexes via
Watson-Crick base pairs and triplexes through Watson-
Crick base pair and Hoogsteen hydrogen bonding, and
PNA-DNA/RNA hybrids exhibit thermal stability higher
than that of analogous DNA:DNA and DNA:RNA com-
plexes.3 Because of this attractive feature and stability
to proteases and nucleases, PNAs are of great interest
in medicinal chemistry,4 with potential for the develop-
ment as gene-targeted drugs (antigene and antisense)
F IGURE 1. Structures of I DNA (RNA), II aegPNA, III
chPNA, and IV cpPNA.
and as reagents in molecular biology and diagnostics.5
However, PNA suffers from drawbacks such as poor
aqueous solubility, cell permeability, and ambiguity in
directional binding of the complementary DNA/RNA, in
both parallel and antiparallel orientations. Attempts to
address these shortcomings have led to several structural
* Fax: 91 20 2589 8320.
(1) (a) Uhlmann, E.; Peyman, A. Chem. Rev. 1990, 90, 366-374. (b)
De Mesmaeker, A.; Haener, R.; Martin, P.; Moser, H. E. Acc. Chem.
Res. 1995, 28, 366-374. (c) Micklefield, J . Curr. Med. Chem. 2001, 8,
1157-1179. (d) Freier, S. M.; Altmann, K.-H. Nucleic Acids Res. 1997,
25, 4429-4443. (e) Dean, N. M. Curr. Opin. Biotechnol. 2001, 12, 622-
625. (f) Braasch, D. A.; Corey, D. R. Chem. Biol. 2001, 8, 1-7. (g) Kool,
E. T. Chem. Rev. 1997, 97, 1473-1487.
(2) (a) Nielsen, P. E.; Egholm, M.; Berg, R. H.; Buchardt, O. Science
1991, 254, 1497-1501. (b) Hyrup, B.; Nielsen, P. E. Bioorg. Med. Chem.
1996, 4, 5-23. (c) Nielsen, P. E. Curr. Opin. Biotechnol. 2001, 12, 16-
21. (c) Petersen, M.; Wengel, J . Trends Biotechnol. 2003, 21, 74. (d)
Nielsen, P. E. Acc. Chem. Res. 1999, 32, 624-630.
(4) (a) Uhlmann, E.; Peyman, A.; Breipohl, G.; Will, D. W. Angew.
Chem., Int. Ed. 1998, 37, 2796-2823. (b) Bennett, C. F. In Applied
Antisense Oligonucleotide Technology; Stein, C. A., Craig, A. M., Eds.;
Wiley-Liss, Inc.: New York, 1998. (c) Braasch, D. A.; Corey, D. R.
Biochemistry 2002, 41, 4503-4510.
(5) (a) Nielsen, P. E.; Egholm, M. Peptide Nucleic Acids (PNA). In
Protocols and Applications; Nielsen, P., E., Egholm, M., Eds.; Horizon
Scientific Press: Norfolk, CT, 1999. (b) Nielsen, P. E. Curr. Opin.
Biotechnol. 2001, 12, 16-20. (c) Nielsen, P. E. In Peptide Nucleic
Acids: Methods and Protocols, Methods in Molecular Biology; Humana
Press: Totowa, NJ , 2004; Vol. 208.
(3) (a) Egholm, M.; Buchardt, O.; Christensen, L.; Behrens, C.;
Freier, S. M.; Driver, D. A.; Berg, R. H.; Kim, S. K.; Norden, B.; Nielsen,
P. E. Nature 1993, 365, 566-568. (b) J ensen, K. K.; Qrum, H.; Nielsen,
P. E.; Norden, B. Biochemistry 1997, 37, 5072-50.
10.1021/jo049442+ CCC: $27.50 © 2004 American Chemical Society
Published on Web 07/30/2004
J . Org. Chem. 2004, 69, 5725-5734
5725

Govindaraju et al.
modifications of PNA.6 The open-chain acyclic PNA
backbone is conformationally flexible, and an effective
way to preorganize the PNA strand for attaining hybrid-
ization competent conformation is to conformationally
rigidify the backbone by introducing bulky substituents
or ring structures into the backbone. Such approaches
providing a conformational lock as in locked nucleic acids
(LNAs) by Wengel et al.7 or freezing the conformations
of six-membered rings as in hexitol nucleic acids (HNAs)8
and carbocyclic nucleic acids such as cylohexanyl (CNA)9
and cyclohexenyl (CeNA)10 nucleic acids by Herdewijn et
al. have resulted in elegant designs of preorganized,
hybridization-competent structures with remarkable suc-
cess in oligonucleotide series. The pioneering work of
Eschenmoser et al.11,12 employing pyranosyl (p-RNA) and
threofuranosyl (TNA) nucleic acids has shed light on the
important conformational differences and the hybridiza-
tion properties of five- and six-membered rings in a
nucleic acid backbone. Deriving inspiration from such
successes, we13 and others14 have employed cyclic PNA
monomers based on a five-membered pyrrolidine or six-
membered pipecolic acid nucleus as part of a PNA
backbone to introduce conformational constraint, with
varied success in terms of DNA/RNA recognition. In one
of the earliest approaches, the six-membered carbocyclic
trans-(1S,2S/1R,2R)-cyclohexyl ring was imposed on the
ethylenediamine segment of the backbone15 but was
ineffective in hybridization to the complementary DNA.
An analysis of the X-ray structural data of the PNA2:
DNA triplex16a and the PNA:DNA duplex16b and NMR
data of the PNA:RNA duplex16c suggested to us that the
dihedral angle â (Table 1) could be a key factor for the
TABLE 1. Dih ed r a l An gles (d eg) in P NA a n d P NA:DNA/
RNA Com p lexes
compound
R
â
γ
δ
ø1
ø2
PNA2:DNA9a
PNA:DNA9b
PNA:RNA9c
chPNA(1S,2R)a
-103
105
170
73
141
67
70
78
79
76
93
1
-175
151
-171
139 -3
84
119
4
128 -63
1.02 -175
chPNA(1R,2S)a -129
66 -78 -119 -0.87
174
165
-165
cpPNA(1S,2R)a
cpPNA(1R,2S)a
84 -24
-84
86
90
-90
0.89
1.2
25 -86
a
Dihedral angles from monomer crystal structures: chPNA,17
cpPNA.19
rational design of a preorganized PNA structure. The
preferred values for â in the PNA2:DNA triplex and the
PNA:RNA duplex are in the range 65-70°, while that
for the PNA:DNA duplex is about ∼140°, suggesting that
it may be possible to impart DNA/RNA binding selectivity
if â is restricted to 65-70° through suitable chemical
modifications.
With such a rationale, we recently reported17 cis-
cyclohexyl PNA (chPNA) in which the axial-equatorial
disposition of cis-1,2 substituents with â in the range
63-66° (opposite in sign for the two enantiomers 1S,2R
and 1R,2S) showed a stereochemistry-dependent prefer-
ential binding of the derived PNA-T oligomers to RNA
as compared to DNA. This was interesting in comparison
to the earlier report on trans-1,2-diaxial-substituted
cyclohexyl system in which the dihedral angle â is 180°
and hence unsuitable for forming hybrids with either
DNA or RNA in agreement with X-ray data. However, a
slightly lower binding affinity for triplex formation was
seen with our chPNA in comparison to unmodified
aegPNA.17 We surmised that despite a favorable â, the
substituted cyclohexyl ring is inherently too rigid as it
gets locked up in either of the chair conformations. A
relatively flexible system would be a cyclopentyl ring in
which the characteristic interconvertible endo-exo puck-
ering that dictates the pseudoaxial/pseudoequatorial
dispositions of substituents18 may allow better torsional
adjustments to attain the necessary hybridization com-
petent conformations. In an attempt to tune the dihedral
angle â in this manner, we recently reported preliminary
studies of cis-(1S,2R/1R,2S)-cyclopentyl PNA-T (cpPNA)-
incorporated homothymine aegPNA chimera that showed
stereochemistry-dependent, sequence-specific triplex for-
mation with complementary DNA and RNA with a high
mismatch discrimination.19 The crystal structures of both
the enantiomeric monomers supported the dihedral angle
rationale. While our work was in progress, Appella et al.20
reported the synthesis and limited studies on monosub-
(6) (a) Ganesh, K. N.; Nielsen, P. E. Curr. Org. Chem. 2000, 4, 1931-
1943. (b) Kumar, V. A. Eur. J . Org. Chem. 2002, 2021-2032.
(7) (a) Petersen, M.; Wengel, J . Trends Biotechnol. 2003, 21, 74-81
(b) Koshkin, A.; Singh, S. K.; Nielsen, P.; Rajwanshi, V. K.; Kumar,
R.; Meldgaard, M.; Olsen, C. E.; Wengel, J . Tetrahedron 1998, 54,
3607-3630. (c) Singh, S. K.; Nielsen, P.; Koshkin, A. A.; Wengel, J .
Chem. Commun. 1998, 455-456. (d) Singh, S. K.; Wengel, J . Chem.
Commun. 1998, 1247-1248.
(8) (a) Lescrinier, E.; Froeyen, M.; Herdewijn, P. Nucleic Acids Res.
2003, 31, 2975-2989. (b) Hendrix, C.; Rosemeyer, H.; Verheggen, I.;
Seela, F.; Arschot, A. V.; Herdewijn, P. Chem. Eur. J . 1997, 3, 110-
120.
(9) Wang, J .; Verbeure, B.; Luyten, I.; Lescrinier, E.; Froeyen, M.;
Hendrix, C.; Rosemeyer, H.; Seela, F.; Arschot, A. V.; Herdewijn, P. J .
Am. Chem. Soc. 2000, 122, 8595-8602.
(10) Mourinsh, Y.; Rosemeyer, H.; Esnouf, R.; Mevedovici, A.; Wang,
J .; Ceulemans, G.; Lescrinier, E.; Hendrix, C.; Busson, R.; Sandra, P.;
Seela, F.; Aerschot, A. V.; Herdewijn, P. Chem. Eur. J . 1999, 5, 2139-
2150.
(11) Pitsch, S.; Wendeborn, S.; Eschenmoser, A. Helv. Chim. Acta.
1993, 76, 2161-2183.
(12) Schoning, K.; Scholz, P.; Guntha, S.; Wu, X.; Krishnamurthy,
R.; Eschenmoser, A. Science 2000, 290, 1347-1351.
(13) (a) Gangamani, B. P.; Kumar, V. A.; Ganesh, K. N. Tetrahedron
1999, 55, 177-192. (b) D’Costa, M.; Kumar, V. A.; Ganesh, K. N. Org.
Lett. 1999, 1, 1513-1516. (c) Lonkar, P. S.; Kumar, V. A. Bioorg. Med.
Chem. Lett. 2004, 14/ 9, 2147-2149. (d) Shirude P. S.; Kumar V. A.;
Ganesh K. N. Tetrahedron Lett. 2004, 45, 3085-3088.
(14) (a) Vilaivan, T.; Khongdeesameor, C.; Harnyauttanokam, P.;
Westwell, M. S.; Lowe, G. Biorg. Med. Chem. Lett. 2000, 10, 2541-
2545. (b) Vilaivan, T.; Khongdeesameor, C.; Harnyauttanokam, P.;
Westwell, M. S.; Lowe, G. Tetrahedron. Lett. 2001, 42, 5533-5536. (c)
Hickman, D. T.; King, P. M.; Cooper, M. A.; Slater, J . M.; Micklefield,
J . Chem. Commun. 2000, 2251-2252. (d) Puschl, A.; Tedeschi, T.;
Nielsen, P. E. Org. Lett. 2000, 2, 4161-4163.
(15) Lagriffoule, P.; Witteng, P.; Ericksson, M.; J ensen, K. K.;
Norden, B.; Buchardt, O.; Nielsen, P. E. Chem. Eur. J . 1997, 3, 912-
919.
(16) (a) Betts, L.; J osey, J . A.; Veal, M.; J ordan, S. R. Science 1995,
270, 1838-1841. (b) Ericksson, M.; Nielsen, P. E. Nat. Struct. Biol.
1996, 3, 410-413. (c) Brown, S. C.; Thomson, S. A.; Veal, J . M.; Davis,
D. J . Science 1994, 265, 777-780.
(17) (a) Govindaraju, T.; Gonnade, R. G.; Bhadbhade, M. M.; Kumar,
V. A.; Ganesh, K. N. Org. Lett. 2003, 5, 3013-3016. (b) Govindaraju,
T.; Kumar, V. A.; Ganesh, K. N. J . Org. Chem. 2004, 69, 1858-1865.
(18) Lambert, J . B.; Papay, J . J .; Khan, S. A.; Kappauf, K. A.;
Magyar, E. S. J . Am. Chem. Soc. 1974, 96, 6112-6118.
(19) Govindaraju, T.; Kumar, V. A.; Ganesh, K. N. Chem. Commun.
2004, 860-861.
5726 J . Org. Chem., Vol. 69, No. 17, 2004

Cyclopentyl PNA
SCHEME 1 a
a
Reagents and conditions: (i) NaN3, NH4Cl, ethanol/water (2:8) reflux, 16 h. (ii) n-Butyric anhydride, dry pyridine, DMAP (cat), rt, 16
h. (iii) P. cepacia (lipase), phosphate buffer, pH 7.2, 1.5 h. (iv) NaOMe in MeOH.
stituted trans-(1S,2S)-cyclopentyl PNA that showed only
weak stabilization of the derived PNA2:DNA triplex over
that of PNA:RNA duplex.
reported ones under identical reaction conditions and in
less reaction time. Further enzymatic treatment of the
mixture of minor (1R,2R)- and major (1S,2S)-butyrate
esters 2 for the second time followed by column purifica-
tion and subsequent methanolysis of the (1S,2S)-2 ester
using NaOMe in methanol gave pure (1S,2S)-3b in 35%
overall yield. The enantiopurity of both (1R,2R)-3a and
(1S,2S)-3b isomers was confirmed by comparison with
the known values of optical rotations reported in the
In this paper we present full results on the chemoen-
zymatic stereospecific synthesis of cis-(1S,2R/1R,2S)-
cpPNA thymine monomers in their enantiomerically pure
form, positionwise incorporation into homothymine
aegPNA oligomers, and DNA/RNA hybridization studies
using UV-Tm measurements, gel electrophoretic shift
assay, and circular dichroism analysis of the cpPNA
hybrids. Electrophoretic gel shift assay shows differential
retardation as a result of hybridization of cpPNA se-
quences with complementary DNA sequences whereas
mismatched hybrids showed weaker and incomplete
complexation. Interesting preferences are seen in DNA/
RNA hybridization that are related to the stereochem-
istry of chiral cpPNA units.
1
literature21 and by H NMR using the chiral chemical
shift reagent Eu(hfc)3. The reduction of the azide function
in (1R,2R)-3a by hydrogenation using Adam’s catalyst24
and in situ t-Boc protection of the resulting amine
function yielded the Boc-protected amino alcohol (1R,2R)-4
(Scheme 2). An attempt to convert alcohol 4 to corre-
sponding mesylate (1R,2R)-5 using mesyl chloride in
pyridine with DMAP as a catalyst was unsuccessful. The
mesylate 5 was obtained by using triethylamine (2.5
equiv) in dry dichloromethane at room temperature in
less than 30 min. The resulting mesylate 5 is very un-
stable to acidic conditions and decomposed on silica gel
during purification and hence was used for further reac-
tion without any purification. The mesylate was treated
with NaN3 in dry DMF to yield the azide (1S,2R)-6, the
reaction accompanied by inversion of configuration at C1.
The azide (1S,2R)-6 was hydrogenated using Adam’s
catalyst to give the amine (1S,2R)-7, which without
further purification was alkylated with ethyl bromoace-
tate in the presence of KF-Celite. This resulted in the
monosubstituted cis-1,2-diamine (1S,2R)-8, which on
acylation with chloroacetyl chloride gave the compound
(1S,2R)-9. The condensation of (1S,2R)-9 with thymine
in the presence of K2CO3 in DMF yielded the desired
(1S,2R)-aminocyclopentyl (thymin-1-yl-acetyl)glycyl PNA
monomer ethyl ester 10a , which on hydrolysis with 0.5
M LiOH in aqueous THF gave the free acid monomer
(1S,2R)-11a . The synthesis of the other enantiomer
(1R,2S)-11b was accomplished starting from alcohol
(1S,2S)-3b following the same steps as described above.
It should be pointed out that our synthetic route involving
resolution steps permits synthesis of both cis enanti-
omers, while the route described for the corresponding
trans analogues13 allows access to only one of the enan-
Resu lts a n d Discu ssion
Syn th esis of cis-(1S,2R)- a n d (1R,2S)-Am in ocyclo-
p en tylglycyl Th ym in e Mon om er s. The synthesis of
title compounds was achieved starting from a mixture
of (+)- and (-)-trans-2-azido cyclopentanoate 2 using
lipases well-known in the literature.21 The butyrate ester
of racemic trans-2-azido cyclopentanol was synthesized
by oxirane ring opening of the commercially available
meso-epoxide 1 with sodium azide22 followed by esterifi-
cation with butyric anhydride (Scheme 1). The racemic
butyrate ester 2 was resolved by enzymatic enantiose-
lective hydrolysis using the lipase from Pseudomonas
cepacia (Amano-PS)23 in sodium phosphate buffer (pH )
7.2), which proceeded with 40% conversion, followed by
chromatography, to afford the optically pure (1R,2R)-
azido alcohol 3a . P. cepacia lipase (Amano-PS) was found
to be very effective for the resolution of racemic-trans-
cyclohexanaote, but lipase from Candida cylinderacea
was totally ineffective as reported.17 We found that the
Amano-PS also gave a slightly better enantiomeric purity
and yield for the azido alcohols 3a and 3b than the
(20) Myers, M. C.; Witschi, M. A.; Larionova, N. V.; Franck, J . M.;
Haynes, R. D.; Hara, T.; Grajkowski, A.; Appella, D. H. Org. Lett. 2003,
5, 2695-2698.
(21) (a) Faber, K.; Honig, H.; Seufer-Wasserthal, P. Tetrahedron Lett.
1988, 29, 1903-1904. (b) Honig, H.; Seufer-Wasserthal, P. J . Chem.
Soc., Perkin Trans. 1 1989, 2341-2345.
(22) Swift, G.; Swern, D. J . Org. Chem. 1967, 32, 511-517.
(23) In an earlier literature procedure,21 this resolution step was
reported for the use of Pseudomonas sp., and it is not clear which
particular enzyme was used.
1
tiomers. All new compounds were characterized by H
and 13C NMR and mass spectral data.
(24) Schaus, S. E.; Larrow, J . F.; J acobsen, E. N. J . Org. Chem. 1997,
62, 4197-4199.
J . Org. Chem, Vol. 69, No. 17, 2004 5727

Govindaraju et al.
SCHEME 2 a
F IGURE 2. UV absorbance (at 260 nm) of mixtures of PNA
16 and the complementary DNA I in the relative molar ratios
of 0:100, 10:90, 20:80, 30:70, 40:60, 50:50, 60:40, 70:30, 80:20,
90:10, and 100:0 (buffer, 10 mM sodium phosphate pH 7.0,
100 mM NaCl, 0.1 mM EDTA).
directed into a pseudoequatorial position. The CD of the
monomers 10a and 10b gave the mirror-image CD sig-
nature as expected for the enantiomeric pairs (Supporting
Information).
Syn th esis of cpP NA Oligom er s. The modified cis-
(1S,2R/1R,2S)-aminocyclopentylglycyl thymine mono-
mers 11a and 11b were incorporated into PNA sequences
using Boc chemistry on L-lysine-derivatized (4-methyl-
benzhydryl)amine (MBHA) resin as reported before,
using HBTU/HOBt/DIEA in DMF as the coupling re-
agent. Various homothymine PNA oligomers (12-20,
Table 1) incorporating modified monomers at the middle
(14 and 17), N-terminus (15 and 18), and C-terminus (13
and 16) and over the entire sequence (homooligomers 19
and 20) were synthesized. The oligomers were cleaved
from the resin using “low-high TFA-TFMSA” proce-
dure28 followed by RP-HPLC purification and character-
ized by mass spectrometry (MALDI-TOF).
UV-Tm Stu d ies of a eg-cpP NA-DNA/RNA Com -
plexes. The hybridization of various cpPNAs with comple-
mentary DNA I sequence was studied by temperature-
dependent UV absorbance experiments. The stoichiometry
for aeg-cpPNA:DNA complexation was established by
J ob’s plot29 of UV absorbance data at 260 nm and found
to be 2:1 (Figure 2). The thermal stabilities (Tm) of aeg-
cpPNA2:DNA triplexes (Table 2, Figure 3) were obtained
for various cpPNAs with complementary DNA I and that
having a single mismatch (DNA II) and poly rA to study
the relative compatibility of the imposed stereochemistry
constraints with deoxyribonucleic and ribonucleic acids.
In general, the PNA2:DNA triplexes were found to be
more stable for cpPNAs containing (1S,2R)- and (1R,2S)-
cyclopentyl thymine units as compared to unmodified
PNA. The cpPNA oligomers 14 and 17 having modifica-
tions in the center of the sequence showed contrasting
binding stabilities as (1S,2R)-cpPNA 14 destabilized the
complex with DNA I (∆Tm ) -23 °C), whereas (1R,2S)-
a
Reagent and conditions. (i) PtO2, dry EtOAc, H2, 35-40 psi,
rt, 3.5 h. (ii) (Boc)2O. (iii) MsCl, dry triethylamine, rt, 0.5 h. (iv)
NaN3, dry DMF, 70 °C, 5 h. (v) PtO2, MeOH, H2, 35-40 psi, rt,
3.5 h. (vi) BrCH2COOEt, KF-Celite, dry CH3CN, 65 °C, 4 h. (vii)
ClCH2COCl, Na2CO3, Dioxan/H2O (1:1),
Thymine, K2CO3 dry DMF, 65 °C, 4 h. (ix) 0.5 M LiOH, aq THF,
0.5 h.
0 °C, 5 min. (viii)
The enantiomeric purity of both alcohols (1R,2R)-3a
and (1S,2S)-3b was confirmed from 1H NMR of corre-
sponding O-acetyl derivatives utilizing the chiral chemi-
cal shift reagent (+)-tris[3-(heptafluoropropylhydroxy-
methylene)-d-camphorato]europium(III) [Eu(hfc)3] (8.0mol
%),25 and the ee for both 3a and 3b was found to be >98%
(for data, see Supporting Information). The torsion angles
â as deduced from the X-ray crystal structures19 for
(1S,2R)-10a and (1R,2S)-10b are -24 and +25° (Sup-
porting Information), respectively, which are much less
than that found in PNA2:DNA and PNA:RNA complexes
and cis-cyclohexyl systems and possess opposite signs,
as expected for the enantiomeric pairs. In solution, the
tertiary amide bond in PNA is known to exist as a
rotameric mixture.9 In the present structures, the amide
bond is trans with carbonyl pointing toward the C-
terminus, similar to that previously seen in cyanuryl
PNA monomer26 and to that seen in the recently reported
crystal structure of the D-lysine-based chiral PNA:DNA
duplex.27 The crystal structure data also shows pseudo-
axial-psuedoequatorial dispositions for the cis-1,2-sub-
stituents in (1S,2R/1R,2S)-cyclopentyl PNA monomers,
with the bulky substituent carrying the nucleobase
(25) Gogte, V. N.; Pandit, V. S.; Natu, A. A.; Nanda, R. K.; Sastry,
M. K. Org. Magn. Res. 1984, 22, 624-630.
(26) Sanjayan, G. J .; Pedireddi, V. R.; Ganesh, K. N. Org. Lett. 2000,
2, 2825-2828.
(28) Christensen, L.; Fitzpatrick, R.; Gildea, B.; Petersen, K. H.;
Hansen, H. F.; Koch, T.; Egholm, M.; Buchardt, O.; Nielsen, P. E.;
Coull, J .; Berg, R. H. J . Peptide Sci. 1995, 3, 175.
(29) (a) J ob, P. Ann. Chim. 1928, 9, 113-203. (b) Cantor, C. R.;
Schimmel, P. R. Biophysical Chemistry; 1980; Part III, pp 624.
(27) Menchise, V.; Simone, G. D.; Tedeschi, T, Corradini, R.; Sforza,
S.; Marchelli, R.; Capasso, D.; Saviano, M.; Pedone, C. Proc. Natl. Acad.
Sci U.S.A. 2003, 100, 12021-12026.
5728 J . Org. Chem., Vol. 69, No. 17, 2004

Cyclopentyl PNA
TABLE 2. UV-Tm of cpP NA:DNA/RNA Tr ip lexesa
entry
PNA
DNA I
DNA II (∆Tm)b
poly rA
1
2
3
4
5
6
7
8
9
aegPNA 12, H-TTTTTTTT-LysNH2
cpPNA 13, H-TTTTTTTtSR-LysNH2
cpPNA 14, H-TTTtSRTTTT-LysNH2
cpPNA 15, H-tSRTTTTTTT-LysNH2
cpPNA 16, H-TTTTTTTtRS-LysNH2
cpPNA 17, H-TTTtRSTTTT-LysNH2
cpPNA 18, H-tRSTTTTTTT-LysNH2
cpPNA 19, H-(tSR)8-LysNH2
45.0
51.0
22.0
44.5
55.0
62.0
48.7
66.6
72.0
34.5 (10.5)
27.8 (23.2)
11.0 (11.0)
26.4 (18.1)
28.0 (27.0)
32.0 (30.0)
27.8 (20.9)
nd
62.0
73.5
76.0
66.0
>85.0
61.0
69.0
>85.0
>85.0
cpPNA 20, H-(tRS)8-LysNH2
nd
a
All values are an average of three independent experiments and accurate to within (0.5°. DNA I, d(CGCAAAAAAAACGC); DNA II,
d(CGCAAAACAAACGC). Buffer: sodium phosphate (10 mM), pH 7.0 with 100 mM NaCl and 0.1 mM EDTA; nd, transition not detected.
Mismatch destabilization.
b
F IGURE 3. Derivative curves of melting profiles of (A) (a) aegPNA 12, (b) cpPNA 13, (c) cpPNA 15, (d) cpPNA 16, (e) cpPNA 18
and (B) (a) cpPNA 19 and (b) cpPNA 20 with DNA I.
cpPNA 17 stabilized the complex (∆T ) +17 °C) in
comparison to the control PNA 12. Thus a remarkable
Tm difference of 40 °C was noticed between the two
enantiomeric PNAs 14 and 17 in binding the comple-
mentary DNA. In the case of aeg-cpPNAs (1S,2R)-13 and
(1R,2S)-16 having single C-terminus modifications, both
stabilized their complexes with DNA (∆T ) +6 and +10
°C, respectively) as compared to aegPNA 12, with a
mutual difference of only +4 °C between themselves. The
N-terminus modified aeg-cpPNAs (1S,2R)-15 and (1R,2S)-
18 showed Tm values almost comparable to the unmodi-
fied PNA but lower compared to respective C-terminus-
modified PNAs.
Significantly, in contrast to singly modified analogues,
the all-modified homooligomeric, homochiral cpPNAs
(1S,2R)-19 and (1R,2S)-20 exhibited high thermal sta-
bilities (∆Tm ) +21 and +27 °C, respectively) compared
to the unmodified PNA 12. The PNA sequence with one
modification in the center with (1S,2R)-stereochemistry,
14, largely destabilized the complex with DNA I. This
could be caused by the introduction of unfavored confor-
mational discontinuity at the modification site. However,
in the fully modified oligomer 19 with the same stereo-
chemistry, the stability is regained perhaps due to a
uniform conformational change over the entire backbone,
without any sharp discontinuities. Such a behavior is not
unusual and has been observed by Nielsen et al in the
case of the pyrrolidine PNA.14d The singly and fully
modified sequences with the other and probably favorable
stereochemistry (1R,2S) in PNA sequences 17 and 20
show better binding affinity toward DNA I, suggesting
that the effect of increased Tm could be additive (1.4 °C
per modification) from a single favorable modification to
fully modified oligomer sequences.
The enhanced thermal stabilities observed for
aeg-cpPNA and cpPNA complexes with DNA is not at the
cost of losing base pairing specificities. This is borne by
data for their complexation with DNA II [dA4CA3] that
carries a single mismatch at the middle position. The
mismatch ∆Tm for control aegPNA 12 was -10.5°, while
that for all cpPNAs were higher, except for (1S,2R)-14,
which was as good. The mismatch destabilization was
highest (-30°) for (1R,2S)-17 with modification at the
center. Other modified PNAs exhibited intermediate
degrees of mismatch destabilization in the range 18-27°.
The lower mismatch tolerance and higher binding affini-
ties to matched complementary DNA I sequence reflects
the sequence specific recognition of cpPNAs. The fully
modified oligomers 19 and 20 did not show any binding
with mismatch DNA II. Complete inhibition of hybrid-
ization with fully modified sequences cpPNA 19 and 20
with DNA II having a mismatch in the center of the
sequence could be a result of mismatched hydrogen
bonding in the center, in turn effecting steric incompat-
ibility of a fully modified sequence for DNA binding. The
open-chain PNAs in other sequences do not impose the
additional steric incompatibility and permit complex
formation with DNA II, albeit reduced stability. In
general, the (1R,2S)-cyclopentyl modified PNAs were
J . Org. Chem, Vol. 69, No. 17, 2004 5729

Govindaraju et al.
F IGURE 4. A: CD spectra of complexes (cpPNA 13)2:DNA I, (cpPNA 14)2:DNA I, (cpPNA 15)2:DNA I, and (cpPNA 19)2:DNA I.
B: CD spectra of complexes (cpPNA 16)2:DNA I, (cpPNA 17)2:DNA I, (cpPNA 18)2:DNA I, and (cpPNA 20)2:DNA I. C: CD spectra
of cpPNA 19 (SS), cpPNA 20 (SS) (10 mM sodium phosphate buffer, pH ) 7, 100 mM NaCl, 0.1 m M EDTA, 20 °C).
better in binding and mismatch differentiation in com-
plexing with complementary DNA.
PNA:DNA triplexes. Figure 4 shows the CD spectra of
complexes of cpPNAs 13-20 with DNA I. The single
stranded cyclopentyl PNAs (13-18) show weak CD bands
in the 250-290 nm region (Supporting Information), and
all cpPNA:DNA triplexes show the expected positive
bands at 258 and 285 nm. The relative ratios of 258/285
nm are different for the cyclopentyl PNA:DNA triplexes
as compared to the control PNA:DNA triplex. These
results imply that the cyclopentyl modifications alter base
stacking patterns in PNA:DNA complexes. Systematic
changes in the CD spectral features also occur in con-
tinuous variation mixing of cpPNA and DNA components
to generate J ob’s plot (Supporting Information) for de-
termining the binding ratio, which corresponded to 2:1
for the cyclopentyl PNA:DNA complexes. Fully modified
single-stranded cpPNAs containing (1S,2R)- or (1R,2S)-
cyclopentyl thymines show weak but distinct CD spectra
(Figure 4C), which are near mirror images of each other
and similar to that of monomers.
In the case of aeg-cpPNA complexation with poly rA,
the presence of either S,R or R,S isomers at C/N terminus
or internally, in general, stabilized the complexes com-
pared to aegPNA. The cpPNA oligomers 14 and 17,
modified at the center showed a reverse selectivity in
binding to RNA with (S,R)-14 > (R,S)-17 by 15° as
compared to binding with DNA I, where (R,S)-17 >
(S,R)-14 by 40°. Unprecedented stabilization was ob-
served for homooligomeric, homochiral (S,R)-18 and
(R,S)-19 cpPNAs binding with poly rA, with no melting
seen up to 85 °C. The overall pattern of Tm values
suggests that cpPNAs have a preference for binding to
RNA over DNA, though this needs to be verified with
mixed-base oligoribonucleotides.
CD Sp ectr oscop ic Stu d ies of cpP NA:DNA Com -
p lexes. Being achiral, aegPNAs show very weak CD
signatures due to the presence of L-lysine at the
C-termius. However, PNA:DNA complexes exhibit char-
acteristic CD signatures due to chirality induced by the
DNA component. It is known that the formation of PNA2:
DNA triplexes30 is accompanied by appearance of positive
CD bands at 258 and 285 nm that are not present in
DNA. The presence of chiral centers in cyclopentyl PNAs
should further influence the CD patterns of the derived
Electr op h or etic Gel Sh ift Assa y a n d Com p etitive
Bin d in g E xp er im en t s. Electrophoretic gel experi-
ment17b,31 was used to establish the binding of modified
PNAs to the complementary DNA I and mismatch DNA
II (Figure 5). The various PNAs were individually mixed
with DNA I in buffer (for conditions, see caption of Figure
5) and subjected to nondenaturing gel electrophoresis at
(31) (a) Sambrook, J .; Fritsch, E. F.; Maniatis, T. In Molecular
Cloning: A Laboratory Manual, 2; Cold Spring Harbor Laboratory
Press: Plainview, NY, 1989.
(30) Kim, S. K.; Nielsen, P. E.; Egholm, M.; Buchardt, O.; Berg, R.
H. J . Am. Chem. Soc. 1993, 115, 6477-6483.
5730 J . Org. Chem., Vol. 69, No. 17, 2004

Cyclopentyl PNA
F IGURE 5. PNA:DNA Complexation. A: lane 1, (aegPNA 12)2 + DNA I; lane 2, ssDNA I; lane 3, sscpPNA 12; lane 4, sscpPNA
14; lane 5, (cpPNA 14)2:DNA I; lane 6, (cpPNA 17)2 + DNA I; lane 7, sscpPNA 17; lane 8, marker bromophenol blue (BPB): B,
lane 1, ssDNA I; lane 2, sscpPNA 13; lane 3, (cpPNA 13)2 + DNA I; lane 4, (cpPNA 15)2 + DNA I; lane 5, (cpPNA 16)2 + DNA
I; lane 6, (cpPNA 18)2 + DNA I; lane 7, BPB. C: Competition binding experiments. Lane 1, DNA I + DNA III + (cpPNA 14)2;
lane 2, DNA I + DNA III + (cpPNA 15)2; lane 3, DNA I + DNA III + (cpPNA 17)2; lane 4, DNA I + (cpPNA 17)2; lane 5, DNA
duplex (I + III); lane 6, ssDNA III; lane 7, ssDNA I. Buffer, 0.5 X TBE, pH 8.0; DNA I ) CGCA8CGC, DNA III ) GCGT8GCG,
BPB ) bromophenol blue.
10 °C, and the bands were visualized on a fluorescent
TLC background. The formation of PNA2:DNA complexes
was accompanied by the disappearance of the band due
to single-strand DNA I and appearance of a lower
migrating band of the complex. The singly modified PNAs
13-18 upon mixing with DNA I migrated about the same
as the unmodified PNA 12:DNA I complex and much
lower than that of DNA I (Figures 5A and 5B), clearly
showing successful formation of cpPNA2:DNA I triplexes.
Under the conditions, the single-stranded PNAs do not
migrate from the well. The gel retardation experiments
for singly modified aeg-cpPNAs with mismatch DNA II
showed weaker and incomplete complexation, as seen by
the presence of faster moving band unbound DNA II
(Supporting Information). Competition binding experi-
ments were carried out by adding aeg-cpPNAs 15 or 17
separately to the DNA duplex (I:III) followed by anneal-
ing and gel electrophoresis. Due to a higher stabilization
of aeg-cpPNA:DNA I complexes compared with the DNA:
DNA duplex, the added PNA binds to the complementary
DNA I in the DNA duplex, concomitantly releasing the
DNA strand III (Figure 5C). This is clearly seen in the
gel electrophoresis by the appearance of lower moving
bands due to the aeg-cpPNA:DNA I complex (Figure 5C,
lanes 2 and 3), a faster moving band due to the released
DNA strand III, and the disappearance of the DNA
(I:III) duplex band (as in lane 5). When the experiment
is performed with aeg-cpPNA14:DNA I complex that has
a lower stability than DNA duplex, the competition
binding is ineffective, as seen by a weak band due to the
PNA:DNA complex and in efficient dissociation of DNA
duplex I:III, which is still persistent (Figure 5C, lane 1).
Thus, the modified aeg-cpPNAs are capable of success-
fully competing for binding with the complementary DNA
I strand in a DNA duplex, just as in unmodified PNA.
Com p a r ison of cis-Cycloh exyl (ch P NA) a n d cis-
Cyclop en tyl (cpP NA) P NAs: Affin ity vs Selectivity.
chPNA and cpPNA are the result of the designs based
on optimizing dihedral angle â to constrain PNA back-
bone for optimal DNA/RNA binding and discrimination
via preorganization-mediated entropy gain. The UV
melting temperatures of the chPNAs with complemen-
tary DNA and RNA suggested a stereochemistry-depend-
ent discrimination, but the thermal stability of resultant
cDNA/RNA complexes was reduced compared to aegPNA.
The cyclohexyl ring although has the dihedral â in the
requisite range of 65-75° has the probable drawback that
the ring gets locked into one of the chair conformations,
which may induce unfavorable conformation in the
resultant PNA oligomer. To improve the binding affinity
and enhance the thermal stability of the complexes
without losing the discrimination and stereospecificity,
cpPNAs were designed on the basis of the fact that the
cyclopentyl ring is less rigid than the cyclohexyl ring, and
unlike the cyclohexyl ring, it has inherent puckering
flexibility with a low-energy barrier for transitions. This
may allow such systems to induce a favorable conforma-
tion toward cRNA binding, inspite of the dihedral angle
â being much less (∼25°) than the cyclohexyl system
(∼65°). Thus, these offer a balance between the complete
rigidity of cyclohexyl systems and the flexibility of open-
chain PNA backbone to fine-tune the dihedral angle â in
modified PNA oligomers. cpPNAs showed a higher bind-
ing affinity to cDNA/RNA with stereochemistry-depend-
ent DNA/RNA discrimination. The binding of cis-cyclo-
pentyl modifications reported here appear to be
J . Org. Chem, Vol. 69, No. 17, 2004 5731

Govindaraju et al.
superior to the trans-cyclopentyl analogues reported
recently.13 The stereoselective binding was reflected in
their lowering Tm values with mismatch cDNA (∆T )
-18 to 30 °C), which is more than that observed with
chPNAs. The relative ease of conformational adjustments
in cis-cyclopentyl system seems to have significant
consequences on the hybridization ability of cpPNA
oligomers despite the less dihedral angle compared to
chPNAs.
of conformational adjustments in a cyclopentyl ring as
compared to rigid locking of cyclohexane systems. The
favorable conformational features of the monomers are
cooperatively transmitted to the oligomeric level. The
results of cis-cpPNAs presented here support the idea of
improving the stability and DNA/RNA binding selectivity
via rational conformational tuning of classical PNA to
achieve preorganization. Further studies on sequence-
dependent and RNA/DNA discriminatory effects of
cpPNA in mixed base cpPNA sequences and thermody-
namic studies to delineate entropic and enthalpic con-
tributions are in progress.
When biochemical recognitions are mediated by steric
fit, higher binding affinity is accompanied by higher
recognition specificities (positive correlation), which is the
case for most enzyme-substrate interactions. A negative
correlation is intrinsic to the nucleation-zipping model,
which governs the sequence recognition between comple-
mentary nucleic acid strands.32 Strategies to improve the
affinity of oligonucleotide analogues along with a corre-
sponding increase in sequence fidelity could involve
enhancing both hydrogen bonding interactions (to have
an enthalpic advantage) and steric fitting (to impart an
entropic advantage). In our previous studies that involved
rational chemical modifications to induce preorganization
of a single-stranded PNA backbone to match that of the
DNA/RNA hybrids, higher affinities are mostly ac-
companied by a higher discrimination of single mis-
matches. In modified analogues with somewhat rigid
structural features, incorrect hydrogen bonding recogni-
tion of DNA sequences with a single-base mismatch
would amplify destabilization effects due to steric inter-
ruption in the zipping mechanism. The observed differ-
ences in DNA affinity between cyclohexyl and cyclopentyl
PNAs may arise from differences in relative contributions
of the above factors, and delineating these may help
achieve optimal fine-tuning of PNA chemical modifica-
tions to achieve balanced affinity/selectivity toward target
sequences.
Exp er im en ta l Section
E n zym a t ic
R esolu t ion
of
R a cem ic
Bu t yr a t e
(1R/S,2R/S)-2 w ith P . cepa cia . To a solution of enzyme P.
cepacia (500 mg) in phosphate buffer (150 mL, pH ) 7.2) was
added racemic butyrate (1R/ S,2R/ S)-2 (10 g). The mixture
was stirred vigorously for 1.5 h, which accomplished 40%
conversion for alcohol (-) (1R,2R)-3a . The enzyme was sepa-
rated by filtration, and the filtrate was extracted into
DCM. The solvent was evaporated, and the column chromato-
graphic separation of the residue afforded chiral pure alcohol
(1R,2R)-3a and optically enriched butyrate (1S,2S)-2. The
optically enriched butyrate (1S,2S)-2 was subjected to second
enzyme hydrolysis as described above to yield the optically
pure butyrate (1S,2S)-2 after column separation. From bu-
tyrate (1S,2S)-2, alcohol (1S,2S)-3b was obtained by metha-
nolysis with a catalytic amount of NaOMe in MeOH in 35%
overall yield.
The enantiomeric purity of 3a and 3b was confirmed by
their optical rotations: (1R,2R)-3a [R]20D -84.0° (lit.13 -78.0°,
c 1.5, CH2Cl2); (1S,2S)-3b; [R]20 +84.0° (lit.13 +84.1°, c 1.6,
D
CH2Cl2).
(1R,2R)-2-(N-t-Boc-Am in o)-cyclop en ta n ol (1R,2R)-4. To
a solution of (1R,2R)-2-azido cyclopentanol (3.5 g, 27.55 mmol)
in dry ethyl acetate (10 mL) placed in a hydrogenation flask
was added di-tert-butyl dicarbonate (7.21 g, 33.06 mmol) and
Adams catalyst (2 mol %). The mixture was hydrogenated in
a Parr apparatus (rt, 35-40 psi, 3.5 h). The catalyst was
filtered, and the solvent in the filtrate was evaporated under
reduced pressure; the residue was purified by column chro-
matography (EtOAc/petroleum ether) to afford a white solid
Con clu sion
To introduce PNA binding selectivity between DNA
and RNA and higher affinity, we have synthesized
cpPNAs incorporating optically pure cis-(1S,2R)- and cis-
(1R,2S)-aminocyclopentylglycyl thymine monomers. These
monomers were built into PNA oligomers by solid-phase
peptide synthesis; UV melting temperatures of the
cpPNAs with complementary and mismatched DNA and
RNA indicated that cpPNAs having a (1S,2R)-cyclopentyl
unit have a higher binding affinity to RNA than DNA,
and PNA oligomers with (1R,2S)-cyclopentyl units showed
relatively higher affinity toward DNA. The stereopref-
erences in cpPNA binding of DNA/RNA is accompanied
by overall enhanced binding affinities compared to aeg-
PNA, unlike chPNA where stereochemistry-dependent
discrimination of DNA/RNA was observed but with lower
Tm values. cis-cpPNAs show high mismatch discrimina-
tion compared to aegPNAs, revealing base-specific rec-
ognition. The torsional angle â in a 1,2-disubstituted cis-
cyclopentyl system is less than that in chPNA but seems
to have significant consequences on the hybridization
ability of cpPNA oligomers. The higher binding of cyclo-
pentyl PNAs is perhaps a consequence of the relative ease
of the alcohol (1R,2R)-4: yield 4.6 g, 83%; mp 87.0 °C; [R]20
D
+21.0° (c 1.0, CH2Cl2); 1H NMR (CHCl3-d, 200 MHz) δH
1.00-1.45 (10H, m, 4-CH, t-Boc), 1.45-1.8 (3H, m, 4-CH,
3-CH2), 1.8-2.15 (2H, m, 5-CH2), 3.4-3.7(1H, m, 2-CH),
3.8-4.0 (1H, m, 1-CH), 4.1-4.4 (1H, bd, -OH), 4.6-5.0 (1H,
bd, carbamate NH); 13C NMR (CHCl3-d, 200 MHz) δC 20.2,
27.9, 29.5, 31.5, 59.5, 78.2, 78.9, 156.5. Anal. Calcd (%) for
C
10H19NO3: C, 59.7; H, 9.45; N, 6.96. Found: C, 59.63; H, 9.81;
N, 6.86. LCMS: 202.05 [M + H].
(1S,2S)-2-(N-t-Boc-Am in o)-cyclopen tan ol (1S,2S)-4. This
was prepared from the azido alcohol (1S,2S)-3b using a
procedure similar to the one described for the alcohol
(1R,2R)-4: mp 87.0 °C; [R]20 -21.0° (c 1.0, CH2Cl2); 1H
D
NMR (CHCl3-d, 200 MHz) δH 1.00-1.45 (10H, m, 4-CH, t-Boc),
1.45-1.8 (3H, m, 4-CH, 3-CH2), 1.8-2.15 (2H, m, 5-CH2),
3.4-3.7 (1H, m, 2-CH), 3.8-4.0 (1H, m, 1-CH), 4.1-4.4
(1H, bd, -OH), 4.6-5.0 (1H, bd, carbamate NH); 13C NMR
(CHCl3-d, 200 MHz): δC 20.2, 27.9, 29.5, 31.5, 59.5, 78.2, 78.9,
156.5. Anal. Calcd (%) for C10H19NO3: C, 59.7; H, 9.45; N, 6.96.
Found: C, 59.67; H, 9.89; N, 6.80. LCMS: 202.05 [M + H].
(1R,2R)-2-(N-t-Boc-Am in o)-cyclop en ta n -1-m eth yl Su l-
fon a te (1R,2R)-5. To a stirred solution of alcohol (1R,2R)-4
(4.3 g, 21.39 mmol) and triethylamine (6.49 g, 64.17 mmol) in
dry CH2Cl2 (40 mL) at 0 °C under nitrogen was added
methanesulfonyl chloride (3.92 g, 34.22 mmol) over a period
of 15 min. The mixture was stirred for another 20 min at room
temperature, and the solvent was evaporated under reduced
(32) Demidov, V. V.; Frank-Kamenetskii, M. D. Trends Biochem.
Sci. 2004, 29, 62-71.
5732 J . Org. Chem., Vol. 69, No. 17, 2004

Cyclopentyl PNA
pressure. The residue was extracted with CH2Cl2, washed
successively with KHSO4 solution, water, and brine, and stored
over Na2SO4. The organic layer was concentrated to afford
mesylate (1R,2R)-5. The yield was 5.6 g, 93%, which was used
without purification for further reaction.
(1S,2S)-2-(N-t-Boc-Am in o)-cyclop en ta n -1-m eth yl Su l-
fon a te (1S,2S)-5. The compound (1S,2S)-5 prepared from the
alcohol (1S,2S)-4 using a procedure similar to the one described
for the mesylate (1R,2R)-5, and the crude product was used
for the further reaction without any purification.
C, 58.74; H, 9.09; N, 9.79. Found: C, 58.85; H, 8.66; N, 9.50.
LCMS: 287.05 [M + H], 187 [M + H - tBoc].
N-[(2S)-t-Boc-Am in ocyclop en t-(1R)-yl]-glycin e Eth yl
Ester (1R,2S)-8. A procedure similar to the one described for
the ethyl ester (1S,2R)-8 afforded (1R,2S)-8 starting from the
amine (1R,2S)-7: [R]D +32.69° (c 1.0, CH2Cl2); 1H NMR
20
(CHCl3-d, 200 MHz); δH 0.62-2.1 (18H, m, 4-CH2, 3-CH2,
5-CH2, t-Boc, ester-CH3), 2.7-3.1 (1H, bd, 2-CH), 3.1-3.4 (2H,
s, -CH2-Cd0), 3.5-3.9 (1H, bd, 1-CH), 3.9-4.2 (2H, q, ester-
CH2), 4.9-5.6 (2H, bd, NH and carbamate-NH); 13C NMR
(CHCl3-d, 200 MHz) δC 13.5, 19.9, 27.8, 29.3, 30.2, 48.6, 52.5,
59.5, 60.1, 78.4, 155.51, 171.2. Anal. Calcd (%) for
(1S ,2R )-2-(N -t -B o c -Am i n o )-1-a z i d o c y c lo p e n t a n e
(1S,2R)-6. A stirred mixture of the mesylate (1R,2R)-5 (5.0 g,
17.92 mmol) and NaN3 (9.3 g, 0.143 mol) in DMF (25 mL)
under nitrogen was heated at 68-70 °C for 5 h. After the
mixture was cooled, the solvent was evaporated under reduced
pressure and the residue was extracted with EtOAc (25 mL x
2) and stored over Na2SO4. The organic layer was removed
under reduced pressure, and the crude product was purified
by column chromatography (EtOAc/petroleum ether) to afford
a white solid of azide (1S,2R)-6: yield (3.65 g, 91.5%); mp
C
14H26N2O4: C, 58.74; H, 9.09; N, 9.79. Found: C, 58.87; H,
8.71; N, 9.72. LCMS; 287.05 [M + H], 187.05 [M + H - tBoc].
N-[(2R)-t-Boc-Am in ocyclop en t a n -(1S)-yl]-N-(ch lor o-
a cetyl)-glycin e Eth yl Ester (1S,2R)-9. To a stirred solution
of the amine (1R,2S)-8 (2.5 g, 8.7 mmol) in 10% Na2CO3 (65
mL) and 1,4-dioxan (65 mL) cooled to 0 °C was added
chloroacetyl chloride (4.93 g, 43.7 mmol) in two additions. After
30 min, the dioxan was removed under reduced pressure and
the residue was extracted into EtOAc (2 × 50 mL) and dried
over Na2SO4. The solvent was evaporated under reduced
pressure, and the residue was purified by column chromatog-
raphy (MeOH/CH2Cl2) affording chloro compound (1S,2R)-9 as
81.0 °C; IR ν (cm-1) (KBr) 3442.7, 3014.53, 2995.24, 2113.84,
1706.88 cm-1; [R]D
+ 116° (c 1.0, CH2Cl2); 1H NMR
20
(CHCl3-d, 200 MHz); δH 1.0-1.45 (11H, m, 4-CH2, t-Boc),
1.65-2.0 (4H, m, 5-CH2, 3-CH2), 3.7-4.1 (2H, m, 1-CH, 2-CH),
4.6-5.0 (1H, bd, carbamate NH); 13C NMR (CHCl3-d, 200
MHz) δC 19.8, 28.2, 28.7, 29.0, 54.7, 64.1, 79.4, 155.3. Anal.
Calcd (%) for C10H18N4O2: C, 53.09; H, 7.96; N, 24.77. Found:
C, 52.55; H, 7.9; N, 24.8. MS (FAB+): 227 (35%) [M + 1], 171
(100%) [M + 1 - N3], 127 (15%) [M + 1 - tBoc].
20
a white solid: yield 2.4 g, 76%; mp 131.0 °C; [R]D +55.0° (c
1
1.0, CH2Cl2); H NMR (CHCl3-d, 500 MHz) δH 1.0-2.4 (18H,
m, 3CH2 cycl, tBoc, ester CH3), 3.5-4.5 (8H, m, 1-CH, 2-CH,
N-CH2, -COCH2Cl, -COOCH2), 5.5-6.0 (1H, bd, carbamate
NH); 13C NMR (CHCl3-d, 500 MHz) δC 13.7, 20.8, 27.8, 28.0,
30.5, 41.1, 46.05, 53.3, 58.8, 61.0, 78.2, 155.5, 167.6, 169.5.
Anal. Calcd (%) for C16H27N2O5Cl: C, 52.89; H, 7.43; N, 7.71;
Cl, 9.77. Found: C, 52.61; H, 7.91; N, 7.58; Cl, 9.72. MS
LCMS: 363.05 [M + H], 263.05 [M + 1 - tBoc].
(1R ,2S )-2-(N -t -B o c -Am i n o )-1-a z i d o c y c lo p e n t a n e
(1R,2S)-6. A procedure similar to the one described for
(1S,2R)-6 was used to prepare (1R,2S)-6 from (1S,2S)-5: mp
81.0 °C; IR ν (cm-1) (KBr) 3442.7, 3014.53, 2995.24, 2113.84,
20
1706.88 cm-1
;
[R]D
-116° (c 1.0, CH2Cl2); 1H NMR
N-[(2S)-t-Boc-Am in ocyclop en t a n -(1R)-yl]-N-(ch lor o-
a cetyl)-glycin e Eth yl Ester (1R,2S)-9. This compound was
prepared from the amine (1S,2R)-8 following the procedure
described for the chloro compound (1S,2R)-9: mp 131.0 °C;
(CHCl3-d, 200 MHz) δH 1.0-1.45 (11H, m, 4-CH2, t-Boc),
1.65-2.0 (4H, m, 5-CH2, 3-CH2), 3.7-4.1 (2H, m, 1-CH, 2-CH),
4.6-5.0 (1H, bd, carbamate NH); 13C NMR (CHCl3-d, 200
MHz) δC 19.8, 28.2, 28.7, 29.0, 54.7, 64.1, 79.4, 155.3. Anal.
Calcd (%) for C10H18N4O2: C, 53.09; H, 7.96; N, 24.77. Found:
C, 52.55; H, 7.98; N, 24.71. MS (FAB+) 227 (35%) [M + 1],
171 (100%) [M + 1 - N3], 127 (15%) [M + 1 - tBoc].
[R]D -55.0° (c 1.0, CH2Cl2); 1H NMR (CHCl3-d, 500 MHz) δH
20
1.0-2.4 (18H, m, 3CH2 cycl, tBoc, ester CH3), 3.5-4.5 (8H, m,
1-CH, 2-CH, N-CH2, -COCH2Cl, -COOCH2), 5.5-6.0 (1H,
bd, carbamate NH); 13C NMR (CHCl3-d, 500 MHz) δC 13.7,
20.8, 27.8, 28.0, 30.5, 41.1, 46.0, 53.3, 58.8, 61.0, 78.2, 155.5,
167.6, 169.5; Anal.Calcd (%) for C16H27N2O5Cl: C, 52.89; H,
7.43; N, 7.71; Cl, 9.77; Found: C, 52.73; H, 7.97; N, 7.61; Cl,
9.75: MS LCMS; 363.05 [M+H], 263.05 [M+1-tBoc].
(1S,2R)-2-(N-t-Boc-Am in o)-1-am in ocyclopen tan e (1S,2R)-
7. To a solution of the azide (1S,2R)-6 (3.0 g, 13.27 mmol) in
methanol (5 mL) taken in hydrogenation flask was added
Adam’s catalyst (2 mol %). The reaction mixture was hydro-
genated in a Parr apparatus for 3.5 h at room temperature
and a H2 pressure of 35-40 psi. The catalyst was filtered off,
and the solvent was removed under reduced pressure to yield
a residue of the amine (1S,2R)-7 as a colorless oil. The yield
was 2.6 g, 98.0%; this compound was used for further reaction
without any purification.
N-[(2R)-t-Boc-Am in ocyclop en t -(1S)-yl]-N-(t h ym in -1-
a cetyl)-glycin e Eth yl Ester (1S,2R)-10a . A mixture of chloro
compound (1S,2R)-9a (2.0 g, 5.52 mmol), thymine (0.70 g, 5.52
mmol), and anhydrous K2CO3 (0.91 g, 6.62 mmol) in dry DMF
(10 mL) under nitrogen was heated with stirring at 65 °C for
4.0 h. After cooling, the solvent was removed under reduced
pressure to leave a residue, which was extracted into DCM (2
× 25 mL) and dried over Na2SO4. The solvent was evaporated,
and the crude compound was purified by column chromatog-
(1R,2S)-2-(N-t-Boc)-1-Am in ocyclopen tan e (1R,2S)-7. The
amine (1R,2S)-7 was obtained on hydrogenation of the azide
(1R,2S)-6 following the same procedure described for the
synthesis of the amine (1S,2R)-7.
raphy (MeOH/DCM) to afford
a white solid of thymine
monomer ethyl ester (1S,2R)-10a : yield 1.8 g, 72.2%; mp
N-[(2R)-t-Boc-Am in ocyclop en t-(1S)-yl]-glycin e Eth yl
Ester (1S,2R)-8. To a stirred mixture of amine (1S,2R)-7 (2.5
g, 12.5 mmol) and freshly prepared KF-Celite (4.35 g, 37.5
mmol) in dry acetonitrile (150 mL) was added ethyl bromo-
acetate (1.87 g, 11.25 mmol) dropwise for 30 min at room
temperature under a nitrogen atmosphere, and the mixture
was heated to 65 °C. After 3.0 h, the Celite was filtered off
and the solvent in the filtrate was evaporated under reduced
pressure, which on column chromatographic purification
(EtOAc) afforded the ethyl ester (1S,2R)-8 as a colorless oil:
211.0-215.0 °C; [R]D +32.64° (c 0.52, CH2Cl2); 1H NMR
20
(CHCl3-d, 500 MHz) δH 1.11-1.2 (3H. t, ester CH3), 1.26-1.4
(10H, tBoc, 4-CH), 1.4-1.53 (1H, m, 4-CH), 1.6-1.7 (1H,m,
5-CH), 1.7-1.78 (1H, m, 3-CH),1.81(3H, s, thymine CH3),
2.02-2.15 (2H, 5-CH), 3.6-4.5 (7H, m, 2-CH, N-CH2, -COCH2,
-COOCH2), 5.0-5.2 (1H, d, 1-CH), 5.4-5.6 (1H, bd, carbamate
NH), 7.04 (1H, s, thymine -CHdC-), 9.4-9.6 (1H, bd,
thymine NH); 13C NMR (CHCl3-d, 500 MHz) δC 12.0, 13.8, 21.3,
28.1, 28.3, 31.1, 46.0, 47.8, 52.4, 58.8, 61.2, 79.8, 110.0, 141.2,
151.1, 155.7, 164.3, 167.3, 169.3. Anal. Calcd (%) for
yield 2.81 g, 78.5%; [R]D -32.69° (c 1.0, CH2Cl2); 1H NMR
20
C
21H32N4O7: C, 55.75; H, 7.07; N, 12.38. Found: C, 55.63; H,
(CHCl3-d, 200 MHz) δH 0.62-2.1 (18H, m, 4-CH2, 3-CH2,
5-CH2, t-Boc, ester-CH3), 2.7-3.1 (1H, bd, 2-CH), 3.1-3.4 (2H,
s, -CH2-Cd0), 3.5-3.9 (1H, bd, 1-CH), 3.9-4.2 (2H, q, ester-
CH2), 4.9-5.6 (2H, bd, NH and carbamate-NH); 13C NMR
(CHCl3-d, 200 MHz) δC 13.5, 19.9, 27.8, 29.3, 30.2, 48.6, 52.5,
59.5, 60.1, 78.4, 155.5, 171.2. Anal. Calcd (%) for C14H26N2O4:
7.45; N, 12.08. MS (FAB+): 453 [M + 1] (42%), 353 (100%) [M
+ 1 - tBoc].
N-[(2S)-t-Boc-Am in ocyclop en t -(1R)-yl]-N-(t h ym in -1-
a cetyl)-glycin e Eth yl Ester (1R,2S)-10b. A procedure simi-
lar to the one described for the monomer (1S,2R)-10 afforded
J . Org. Chem, Vol. 69, No. 17, 2004 5733

Govindaraju et al.
zhydryl) amine (MBHA) resin (initial loading 0.25 mequiv g-1
with HBTU/HOBt/DIEA in DMF/DMSO as a coupling reagent.
The PNA oligomers were cleaved from the resin with TFMSA.
The oligomers were purified by RP HPLC (C18 column) and
characterized by MALDI-TOF mass spectrometry. The overall
yields of the raw products were 35-65%. The normal PNAs
were prepared as described in the literature.
20
(1R,2S)-10b from (1R,2S)-9b: mp 211.0-215.0 °C. [R]D
)
-32.64° (c 0.52, CH2Cl2); 1H NMR (CHCl3-d, 500 MHz) δH
1.11-1.2 (3H. t, ester CH3), 1.26-1.4 (10H, tBoc, 4-CH),
1.4-1.53 (1H, m, 4-CH), 1.6-1.7 (1H,m, 5-CH), 1.7-1.78 (1H,
m, 3-CH),1.81(3H, s, thymine CH3), 2.02-2.15 (2H, 5-CH),
3.6-4.5 (7H, m, 2-CH, N-CH2, -COCH2, -COOCH2),
5.0-5.2 (1H, d, 1-CH), 5.4-5.6 (1H, bd, carbamate NH), 7.04
(1H, s, thymine -CHdC-), 9.4-9.6 (1H, bd, thymine NH);
13C NMR (CHCl3-d, 500 MHz) δC 12.0, 13.8, 21.3, 28.1, 28.3,
31.1, 46.0, 47.8, 52.4, 58.8, 61.2, 79.8, 110.0, 141.2, 151.1, 155.7,
164.3, 167.3, 169.3. Anal.Calcd (%) for C21H32N4O7: C, 55.75;
H, 7.07; N, 12.38. Found: C, 55.69; H, 7.37; N, 12.16. MS
(FAB+): 453 [M + 1] (42%), 353 (100%) [M + 1 - tBoc].
N-[(2R)-t-Boc-Am in ocyclop en t -(1S)-yl]-N-(t h ym in -1-
a cetyl)-glycin e (1S,2R)-11a . To the monomer ester (1S,2R)-
10a (1.0 g, 2.283 mmol) suspended in THF (10 mL) was added
a solution of 0.5 M LiOH (10 mL, 5.1 mmol), and the mixture
was stirred for 30 min. The mixture was washed with EtOAc
(2 × 10 mL). The aqueous layer was acidified to pH 3 and
extracted with EtOAc (4 × 20 mL). The EtOAc layer was dried
over sodium sulfate and evaporated under reduced pressure
to afford monomer (1S,2R)-11a as a white solid: yield 0.92 g,
MALDI-TOF. PNA 12: Calcd for C94H126N34O34, 2274.0;
found, 2279.0 [M + 5H]+. PNA 13: Calcd for C97H130N34O34
,
2314.0; found, 2320.0 [M + 6H]+. PNA 14: Calcd for
C
97H130N34O34, 2314.0; found, 2321.0 [M + 7H]+. PNA 15:
Calcd for C97H130N34O34, 2314.0; found, 2321.0 [M + 7H]+. PNA
16: Calcd for C97H130N34O34, 2314.0; found, 2316.0 [M + 2H]+.
PNA 17: Calcd for C97H130N34O34, 2314.0; found, 2321.0 [M +
7H]+. PNA 18: Calcd for C97H130N34O34, 2314.0; found, 2321.0
[M + 7H]+. PNA 19: Calcd for C118H158N34O34, 2594.0; found,
2596.0 [M + 2H]+. PNA 20: Calcd for C118H158N34O34, 2594.0;
found 2596.0 [M + 2H]+.
UV-Tm Mea su r em en ts. The concentration of PNA and
DNA was calculated on the basis of absorbance from the molar
extinction coefficients of the corresponding nucleobases. The
complexes were prepared in 10 mM sodium phosphate buffer,
pH 7.0, containing NaCl (100 mM) and EDTA (0.1 mM) and
were annealed by keeping the samples at 95 °C for 5 min
followed by slow cooling to room temperature. The samples
were cooled by keeping at 4 °C for overnight. Absorbance
versus temperature profiles were obtained by monitoring at
260 nm with Perkin-Elmer Lambda 35 UV-vis spectropho-
tometer scanning from 5 to 85/90 °C at a ramp rate of 0.2 °C
per minute. The data were processed using Microcal Origin
5.0 and Tm values derived from the derivative curves.
20
1
97.8%; mp 139.0-141.0 °C; [R]D -17.0° (c 1.0, CH2Cl2); H
NMR (DMSO-d6, 300 MHz) δΗ 1.0-2.0 (18 Η, thymine CH3,
Boc, 3CH2), 3.5-3.75 (m, 1H), 3.85-4.25 (m, 3H), 4.3-4.7 (m,
1H), 4.8-5.1 (bd, 1H), 6.9-7.15 (bd, 1H), 7.29 (s, 1H), 11.28
(s, 1H); 13C NMR (DMSO-d6, 200 MHz) δC 12.0, 21.8, 28.6,
28.31, 30.0, 38.4, 45.8, 48.1, 50.54, 59.0, 78.5, 108.4, 141.9,
151.4, 156.0, 164.6, 167.9, 171.2. Anal. Calcd (%) for
C
19H28N4O7: C, 53.77; H, 6.60; N, 13.20. Found: C, 53.57; H,
6.91; N, 13.11. MS (FAB+): 425 [M+1] (6%), 325 (100%) [M +
1 - tBoc].
Electr op h or etic Gel Mobility Sh ift Assa y. The PNAs
(12-18, Table 1) were individually mixed with DNA (I or II)
in a 2:1 ratio (PNA strand, 0.4 mM, and DNA II, 0.2 mM or I)
in water. The samples were lyophilized to dryness and
resuspended in sodium phosphate buffer (10 mM, pH 7.0,
10µL) containing EDTA (0.1 mM). The samples were annealed
by heating to 85 °C for 5 min followed by slow cooling to rt
and refrigeration at 4 °C overnight. To this was added 10 µL
of 40% sucrose in TBE buffer pH 8.0, and the sample was
loaded on the gel. Bromophenol blue (BPB) was used as the
tracer dye separately in an adjacent well. Gel electrophoresis
was performed on a 15% nondenaturing polyacrylamide gel
(acrylamide/bisacrylamide 29:1) at constant power supply of
200 V and 10 mA, until the BPB migrated to three-fourths of
the gel length. During electrophoresis, the temperature was
maintained at 10 °C. The spots were visualized through UV
shadowing by illuminating the gel placed on a fluorescent silica
gel plate, F254, using UV light.
N-[(2S)-t-Boc-Am in ocyclop en t -(1R)-yl]-N-(t h ym in -1-
a cetyl)-glycin e (1R,2S)-11b. To the monomer ester (1S,2R)-
10b (1.0 g, 2.283 mmol) suspended in THF (10 mL) was added
a solution of 0.5 M LiOH (10 mL, 5.1 mmol), and the mixture
was stirred for 30 min. The mixture was washed with EtOAc
(2 × 10 mL). The aqueous layer was acidified to pH 3 and
extracted with EtOAc (4 × 20 mL). The EtOAc layer was dried
over sodium sulfate and evaporated under reduced pressure
to afford monomer (1S,2R)-11b as a white solid: yield 0.92 g,
20
1
97.8%; mp 139.0-141.0 °C; [R]D +17.0° (c 1.0, CH2Cl2); H
NMR (DMSO-d6, 300 MHz) δΗ 1.0-2.0 (18 Η, thymine CH3,
Boc, 3CH2), 3.5-3.75 (m, 1H), 3.85-4.25 (m, 3H), 4.3-4.7 (m,
1H), 4.8-5.1 (bd, 1H), 6.9-7.15 (bd, 1H), 7.29 (s, 1H), 11.28
(s, 1H); 13C NMR (DMSO-d6, 200 MHz) δC 12.0, 21.8, 28.6, 28.3,
30.0, 38.4, 45.8, 48.1, 50.5, 59.0, 78.5, 108.4, 141.9, 151.4, 156.0,
164.6, 167.9, 171.2. Anal. Calcd (%) for C19H28N4O7: C, 53.77;
H, 6.60; N, 13.20. Found: C, 53.49; H, 6.93; N, 13.17. MS
(FAB+): 425 [M + 1] (6%), 325 (100%) [M + 1 - tBoc].
En a n tiom er ic P u r ity fr om 1H NMR Usin g a Ch ir a l
Ch em ica l Sh ift Rea gen t. The enantiomeric purity of alcohols
(1R,2R)-3a and (1S,2S)-3b was confirmed from 1H NMR
measurement of their acetyl derivatives utilizing a chiral
chemical shift reagent, (+)-tris[3-(heptafluoropropylhydroxy-
methylene)-d-camphorato]europium (III) [Eu(hfc)3] (8.0 mol %).
Examination of the 1H NMR spectrum of the acetyl derivatives
of racemic (1R/ S,2R/ S)-3 shows the splitting in the methyl
carbon signal into a clear doublet. But in the corresponding
spectra of acetyl derivatives of (1R,2R)-3a and (1S,2S)-3b, the
above-mentioned signal was a singlet and corresponded to a
respective signal in the spectrum of their racemate. Integration
of the methyl signal (doublet) in the racemate and that of the
pure (1R,2R)-3a and (1S,2S)-3b (singlet) shows that the
enatiomeric excess will be >98% for both isomers (see Sup-
porting Information)
Ack n ow led gm en t. T.G. thanks CSIR, New Delhi,
for the award of a research fellowship.
Su p p or tin g In for m a tion Ava ila ble: General experimen-
tal procedures; enantiopurity of alcohols 3a and 3b; 1H NMR,
13C NMR, and mass spectra of compounds 4-6 and 8-11; IR
spectrum of (1S,2R)-6, CD curves for cpPNA monomers
(1S,2R)-10a and (1R,2S)-10b; HPLC, MALDI-TOF, and char-
acterization data table for PNAs 12-20; CD spectra of PNA
12, DNA I, single-strand cpPNAs, and complex PNA 12:DNA
I; CD J ob’s plots for PNAs 14 and 16; UV J ob’s plot for PNA
14; UV-melting curves for complexes of PNAs 12-20 with
DNA I [d(CGCA8CGC)], DNA II [d(CGCA4CA3CGC], and poly
rA; gel shift assay for the complexation of PNAs 13-15, 17,
and 18 with DNA II; X-ray crystal structures of cpPNA
monomer esters 10a , and 10b. This material is available free
of charge via the Internet at http://pubs.acs.org.
Syn th esis of P NA-Oligom er s In cor por atin g cis-(1S,2R)-
an d (1R,2S)-Am in ocyclopen tyl P NA Mon om er s. The modi-
fied PNA monomers were built into PNA oligomers using
standard procedure on a L-lysine-derivatized (4-methylben-
J O049442+
5734 J . Org. Chem., Vol. 69, No. 17, 2004